Joint-space Kalman-based compensation for computation delays in leader-follower coordination of two manipulators

A predictive scheme to compensate for computation delays is developed. Since the research is heavily experimental a leader-follower coordination scheme is adopted because of its computational efficiency. The developed coordination and prediction algorithms are experimentally tested using a hierarchical multiprocessor architecture and two PUMA robot arms. The computation of desired follower arm joint positions consumes one sampling interval. Another sampling interval is lost due to the fact that a desired follower arm position has to be computed before it can be assumed. This is where Kalman-based prediction is used to compensate for these delays